Tecoma

Authority

Gentry, Alwyn H. 1992. Bignoniaceae--part II (Tribe Tecomeae). Fl. Neotrop. Monogr. 25: 1-370. (Published by NYBG Press)

Family

Bignoniaceae

Scientific Name

Tecoma

Synonyms

Stenolobium, Tecomaria, Kokoschkinia

Description

Genus Description - Shrubs or small trees (rarely subscandent in T. capensis). Leaves simple, 3-foliolate or imparipinnately compound, the leaflets serrate. Inflorescence a terminal raceme or racemose panicle. Flowers with the calyx cupular, with five shallow, often apiculate deltoid lobes; corolla yellow or orange-red, tubular-campanulate to narrowly tubular-salverform, glabrous outside, the stamens exserted or included, the anthers divaricate to basally divergent, glabrous or pilose; pollen grains single, oblong, 3-colporate, the exine microreticulate; ovary narrowly cylindric, more or less lepidote, the ovules 2-seriate in each locule; disk cupular-pulvinate to pulvinate-cylindric. Fruit a linear capsule, somewhat compressed parallel to the septum but dehiscing perpendicular to it, the valves smooth, more or less glabrous; seeds thin, bialate, the wings hyaline-membranaceous, sharply demarcated from the seed body.

Discussion

Tecoma is a taxonomically difficult group with poorly demarcated species mostly differentiated by variable and often complexly overlapping vegetative characters. There are two basic species groups, one with narrowly tubular orange or red-orange hummingbird-pollinated flowers and one with campanulate yellow bee-pollinated flowers. The yellow-flowered group consists of one wide-ranging polymorphic species, T. stans, and three geographically restricted segregates—one with simple leaves in western Ecuador, one with 3-foliolate leaves in northwestern Peru and southwestern Ecuador, and one with slightly more obtuse leaflets at higher altitudes in the Andes of Peru and Bolivia. The hummingbird-pollinated group is restricted to the central Andes, from Peru to northern Chile and Argentina, and appears to be undergoing active evolutionary divergence with more or less recognizable but confusingly overlapping morphotypes isolated in different inter-Andean valleys. In attempting a taxonomic treatment of these taxa, I have been guided almost as much by geography as morphology. As here delimited, each hummingbird-pollinated species has a unique and allopatric range, even though several species thus include a few aberrant collections which exhibit morphological features that usually characterize a different species. For example, collections with long-exserted anthers from the Apurimac Valley are referred to local T. arequipensis (which they match vegetatively) rather than to T. cochabambensis, and a single collection with small black-drying acuminate and sharply serrate leaflets from the Marañón drainage is referred to local T. rosifolia (which tends to have a similarly curved corolla tube) rather than to Bolivian-Argentinian T. tenuiflora. I have attempted to include such variants in the key, even though this means that some key characters are somewhat overlapping. The extent to which their existence compromises specific recognition of the various allopatric populations remains an open question, especially since all species tested, including members of the two different pollination guilds, appear to be interfertile (Gentry, 1990: 120). I here follow van Steenis (1977) in reducing African Tecomaria to Tecoma, contrary to my earlier opinion (Gentry, 1977). Although established custom supports the recognition of two genera, characters for differentiating them are vanishingly few. The only separating character of potential generic significance is fusion of the apical portion of the thus divergent-based anthers with the filament apex in Tecomaria, but not Tecoma. Although the neotropical Tecoma species have uniformly divaricate anther thecae, other bignon genera that similarly include both bird- and bee-pollinated species often have the thecae varying from divaricate in the bee-pollinated taxa to pendulous and more or less parallel in the bird-pollinated ones. Thus the change from divaricate to pendulous thecae would not seem to be a very fundamental one, and the subsequent partial fusion of the apical part of the Tecomaria thecae does not seem, on balance, to warrant generic separation. Tecoma and Tecomaria are interfertile (see Gentry, 1990), also supporting their merger. Indeed the similarities of Tecomaria with Tecoma are so great that Seemann (1863) argued that the South African species must not be truly native but an introduction from the New World, and treated the hummingbird-pollinated New World Tecoma species as belonging in Tecomaria rather than as congeneric with the bee-pollinated neotropical species. Although it is now obvious that Seemann’s (1863) third-hand observations were flawed and T. capensis is native only to southern Africa, modem analysis of angiosperm biogeography in the light of plate tectonics (e.g., Raven & Axelrod, 1974) no longer makes such a distribution implausible.

Distribution

A genus of 14 species, 2 in Africa and 12 in the Neotropics. The neotropical species range from extreme southern United States through Central America and the Antilles south through Andean South America to northern Argentina. One of the African species is widely cultivated in the Neotropics and both are included in this treatment.

Africa|